Bipolar disorder and subsequent Parkinson's disease: a meta-analysis of cohort studies.

BACKGROUND

The association between bipolar disorder (BD) and Parkinson's disease (PD) has been examined in several cohort studies, but existing evidence is limited by potential confounding and outcome misclassification. This meta-analysis aims to systematically evaluate this association while appraising the certainty of the evidence.

METHODS

A comprehensive literature search was conducted in PubMed, Web of Science, Embase and the Cochrane Library from database inception to April 2026. Cohort studies reporting PD risk in individuals with BD were eligible for inclusion. The outcome was the incidence of PD, with hazard ratios (HRs) and 95% confidence intervals (95% CIs) as effect measures. Random-effects models were employed and 95% prediction intervals (95% PIs) were calculated to reflect the expected range of true effects across studies. Effect estimates were extracted as reported, with varying degrees of adjustment across studies.

RESULTS

Six cohort studies fulfilled the predefined inclusion criteria. Pooled analysis demonstrated extreme heterogeneity (I 2 = 92.7%, P < 0.001), with an HR of 3.65 (95% CI 2.16-6.17, P < 0.001, 95% PI, 0.67-20.00). Subgroup analyses were exploratory. Any observed differences by geographic region likely reflect variation in healthcare systems rather than biological effects. These findings are drawn from a limited number of observational studies with high risk of confounding and should be interpreted with caution.

CONCLUSION

This meta-analysis suggests an association between BD and subsequent PD diagnosis, with very low certainty of evidence. The pooled estimate should not be interpreted as representing a single underlying effect, and a causal interpretation is not warranted due to uncontrolled medication exposure, outcome misclassification, and extreme heterogeneity.